A large study investigating whether low-cost devices can help people stick to their medication regimens found scant evidence that any of the devices were associated with improved adherence. The study did not focus on people with HIV, but its findings raise questions about scalable methods to improve antiretroviral adherence nevertheless.

Publishing their findings in JAMA Internal Medicine, researchers in the REMIND study conducted a four-arm, block-randomized clinical trial of 53,480 enrollees of the national pharmacy benefit manager CVS Caremark who had commercial insurance. The participants were 18 to 64 years old and taking one to three oral medications for long-term use and were all suboptimally adherent to those drugs, taking an estimated 30 to 80 percent of prescribed doses during the 12 months prior to entering the study.

The participants were divided into groups based on the types of medications they were using when they entered the study: either medications for chronic diseases (e.g., cardiovascular disease) or antidepressants. The participants were further divided into subgroups based on whether they were prescribed to take one pill per day or more than one pill per day.

Members of the respective subgroups were then randomized to receive one of three interventions or to be in the control group. They received in the mail one of the following: 1) a pill bottle with an affixed strip with toggles that could be slid after taking each dose; 2) a pill bottle cap with a digital timer displaying the time elapsed since the last dose was taken; 3) a standard plastic pillbox with one compartment for each day of the week; or 4) no device (the control group).

Of the 37,532 people in the chronic disease group, 6,242 were in the control group, and a respective 12,482, 12,478 and 6,330 were in the groups receiving the day-of-the-week pillbox, digital-timer pillbox cap, and pill bottle with strip and toggles. Of the 15,948 people in the antidepressant group, a respective 2,638, 5,273, 5,271 and 2,766 were randomized into the four intervention groups.

The study was conducted between February 2013 and March 2015. The average age of the participants was 45. Fifty-six percent were women.

The researchers found no association between the intervention the participants received and the proportion of them with optimal medication adherence (at least 80 percent of doses taken, based on pharmacy refill data).

Directly comparing individual interventions with other interventions, the researchers found that those who received the standard day-by-day pillbox tended to have higher adherence than those who received either of the other two interventions. However, this benefit was small and inconsistently statistically significant, meaning in many cases the apparent benefit could have been driven by chance.

Looking at the study’s limitations, the researchers speculated that low or inconsistent use of the medication-reminder devices might have reduced their overall benefit. Additionally, because the mailing of the devices was not coordinated with medication refill times, study participants may have found it awkward to coordinate beginning to use the devices with the pills they had available at that moment.

The researchers found that adherence rates improved in the subgroups randomized to the control group, with 12 to 18 percent of them becoming optimally adherent without any intervention during the study. Such a natural shift in adherence patterns raises the question of whether the study authors would have done better to recruit people who were not necessarily poorly adherent to their medications at the study’s outset but who were predicted to adhere poorly in the near future.

To read the study abstract, click here.